Your data science team has finished a 12-month churn analysis that processed more than 300 GB of clickstream and transactional data. Deliverables include an interactive notebook for peer review, a drill-down dashboard for product managers, and a five-slide executive leadership deck that must let C-level stakeholders decide whether to fund additional retention initiatives. Which data selection is most appropriate to include in the leadership deck?
A complete ranking of all engineered features by their SHAP importance values to demonstrate which factors were most influential in the model's predictions.
A presentation of the final model's performance metrics, including its ROC curve, AUC score, and precision-recall curve, to establish the model's statistical validity.
A single-page summary that aggregates quarterly churn rate versus target, shows the net revenue impact, and highlights the three highest-impact drivers in a concise chart or table.
A detailed week-over-week breakdown of churn rates across all 25 customer segments identified during the analysis to show the granularity of the findings.
For a C-level audience making a funding decision, the most effective communication focuses on high-level business outcomes. The correct option provides a concise summary of performance against targets (quarterly churn), financial implications (net revenue impact), and actionable insights (top three drivers). While model performance metrics (like AUC and ROC curves), granular segmentation data, and feature importance rankings are crucial for technical validation and peer review, they are too detailed for an executive summary. Executives are primarily concerned with the 'so what?' - the business impact and the basis for a decision - not the underlying statistical or engineering details of the model.
Ask Bash
Bash is our AI bot, trained to help you pass your exam. AI Generated Content may display inaccurate information, always double-check anything important.
Why is it important to focus on high-level business outcomes in presentations for C-level stakeholders?
Open an interactive chat with Bash
What are ROC curves and AUC scores, and why aren't they suitable for an executive presentation?
Open an interactive chat with Bash
How do SHAP values contribute to model explainability, and why are they less relevant in executive summaries?